Does metformin extend life? Diabetics on it outlived non-diabetics in one cohort

Metformin — a cheap, decades-old diabetes drug — has become the most talked-about candidate for a pill that slows human ageing. The excitement traces largely to one striking finding: people with type 2 diabetes started on metformin appeared to outlive matched people without diabetes at all [s1]. If a drug can do that in patients who start with a disease, the reasoning goes, it might extend healthy life in everyone. It is a genuine signal worth testing. It is also, as of now, entirely observational — and the trial designed to settle it has still not run.

The evidence

The influential study used UK primary-care records to follow 78,241 people with type 2 diabetes started on metformin, 12,222 started on a sulphonylurea, and 90,463 matched people without diabetes, over a censored total of 503,384 person-years, during which 7,498 died [s1]. The comparison is the part that travelled. Taking survival on metformin as the reference, adjusted median survival was 15% lower in the matched non-diabetic controls (survival time ratio 0.85; 95% confidence interval 0.81 to 0.90) and 38% lower in the diabetic patients on a sulphonylurea (0.62; 95% CI 0.58 to 0.66) [s1]. People with a serious metabolic disease, on this analysis, lived slightly longer than people without it — provided the drug they took was metformin.

A later systematic review pooled this literature. Across 53 eligible studies, diabetics taking metformin had significantly lower all-cause mortality than non-diabetics (hazard ratio 0.93; 95% CI 0.88 to 0.99), and lower mortality than diabetics on other therapies (hazard ratio 0.72; 95% CI 0.65 to 0.80 versus non-metformin treatments) [s3]. Metformin users also showed modestly lower cancer rates than non-diabetics (rate ratio 0.94; 95% CI 0.92 to 0.97) [s3]. The consistency is why researchers proposed testing metformin not as a diabetes drug but as a geroprotector — one that might delay several age-related diseases at once [s2].

The TAME trial — the plan, not the result

That proposal became TAME, short for Targeting Aging with Metformin: a randomised trial designed to test whether metformin delays the onset of a composite of age-related conditions and death in older adults who do not have diabetes [s2]. Its scientific logic is that ageing itself is the shared risk factor behind heart disease, cancer and dementia, so a drug that touches the underlying biology could move all of them together [s2]. TAME was also conceived as a regulatory proof of concept — an attempt to show that "ageing" can be a trial endpoint at all.

The crucial caveat is that TAME is a design, not a dataset. As a large, multi-year prevention trial in non-diabetic older adults, it has faced persistent funding hurdles, and no published TAME result exists to confirm or refute the observational signal [s2]. Everything currently claimed about metformin and human lifespan rests on the cohort and meta-analytic evidence above.

The interpretation trap

Observational comparisons of drug users are vulnerable to exactly the biases that would manufacture a false longevity effect. People prescribed metformin first tend to be healthier and earlier in disease than those escalated to sulphonylureas or insulin, so some of the apparent benefit is confounding by indication, not pharmacology [s1][s3]. The review's authors flagged baseline differences between groups that statistical adjustment can reduce but not erase [s3]. And the eye-catching "diabetics outlive non-diabetics" result is a comparison of adjusted median survival within one records database, not a controlled experiment [s1] — the kind of finding that trials exist precisely to check.

There is also a mechanistic wrinkle: in some studies, metformin has blunted the fitness gains from exercise, a reminder that a drug altering metabolism broadly may carry trade-offs a mortality curve does not show.

The honest use

Metformin is a well-understood, inexpensive diabetes medicine with a strong safety record in the people it is approved for, and the hypothesis that it slows aspects of ageing is serious enough to deserve a proper trial [s2]. What the evidence does not yet support is taking it for longevity if you do not have diabetes. The human data are observational, the standout finding is a database comparison open to confounding, and the trial built to answer the question has not delivered a verdict [s1][s2][s3]. Until it does, metformin-as-longevity-drug is a promising hypothesis wearing the costume of a proven one.

This article is informational and is not medical advice. Do not start or stop a prescription medicine without a clinician.

Sources

1. [s1] Can people with type 2 diabetes live longer than those without? A comparison of mortality in people initiated with metformin or sulphonylurea monotherapy and matched, non-diabetic controls. Diabetes, Obesity and Metabolism, 2014. https://doi.org/10.1111/dom.12354
2. [s2] Metformin as a Tool to Target Aging. Cell Metabolism, 2016. https://doi.org/10.1016/j.cmet.2016.05.011
3. [s3] Metformin reduces all-cause mortality and diseases of ageing independent of its effect on diabetes control: A systematic review and meta-analysis. Ageing Research Reviews, 2017. https://doi.org/10.1016/j.arr.2017.08.003